Title :
Observer-based closed-loop control for the glucose-insulin system: Local Input-to-State Stability with respect to unknown meal disturbances
Author :
Palumbo, P. ; Pepe, P. ; Panunzi, S. ; De Gaetano, Andrea
Author_Institution :
Ist. di Analisi dei Sist. e Inf., Italian Nat. Res. Res. Council, Rome, Italy
Abstract :
Closed-loop glucose control schemes, usually based on intravenous/subcutaneous insulin administration, need to cope with the problem of exogenous glucose intake (e.g. a meal), a disturbance hard to anticipate in timing, in amount and in the rate of effective absorption of the nutrient. In this note, an intravenous feedback control law is considered, based on the use of a Delay Differential Equation (DDE) model of the glucose-insulin system, with the external meal treated as a completely unknown disturbance. It is shown that the closed-loop system satisfies the local Input-to-State Stability (ISS) property with respect to the unknown disturbance. The equivalence between asymptotic stability and local input-to-state stability for retarded nonlinear systems is proven. Simulations show the efficacy of the control algorithm with respect to a standard plasma glucose appearance rate profile following a meal, taken from the literature.
Keywords :
asymptotic stability; closed loop systems; delays; differential equations; feedback; medical control systems; nonlinear control systems; observers; sugar; DDE; ISS; asymptotic stability; closed-loop glucose control schemes; delay differential equation model; exogenous glucose intake; glucose-insulin system; input-to-state stability property; intravenous feedback control law; intravenous-subcutaneous insulin administration; local input-to-state stability; observer-based closed-loop control; plasma glucose appearance rate profile; retarded nonlinear systems; unknown meal disturbances; Asymptotic stability; Diabetes; Insulin; Mathematical model; Plasmas; Stability analysis; Sugar; Glucose-Insulin System; Input-to-State Stability; Retarded Systems;
Conference_Titel :
American Control Conference (ACC), 2013
Conference_Location :
Washington, DC
Print_ISBN :
978-1-4799-0177-7
DOI :
10.1109/ACC.2013.6580089